Transform Education with LearnLens

AI-powered assessment generation, knowledge tracing, and personalized recommendations that close the loop between assessment and mastery.

Smart Assessment

Auto-generate questions with knowledge concept tagging

Knowledge Tracing

Track mastery with Bayesian Knowledge Tracing

Smart Recommendations

Personalized learning paths with RAG-powered insights

Our Features

A comprehensive suite of AI-powered tools designed to revolutionize educational assessment and personalized learning.

Assessment Generator

Automatically generate high-quality assessments from lecture materials, slides, and transcripts using multi-LLM architecture.

  • Multi-format question generation
  • Knowledge concept auto-tagging
  • Difficulty calibration
  • Bloom's taxonomy alignment

Knowledge Tracer

Track student mastery in real-time using Bayesian Knowledge Tracing and predict learning outcomes.

  • Real-time mastery estimation
  • Predictive learning analytics
  • Skill dependency mapping
  • Intervention recommendations

Student Recommender

Personalized remediation with semantic search. Connect struggling students to the exact resources they need.

  • Vector-based content matching
  • Personalized learning paths
  • RAG-powered explanations
  • Adaptive practice problems

Teacher Dashboard

Cohort-level analytics with error clustering and confusion matrix visualization for targeted interventions.

  • Error pattern clustering
  • Confidence vs. accuracy heatmaps
  • Misconception detection
  • Curriculum gap analysis

Multi-Cloud Architecture

Built on a robust multi-cloud foundation with Azure, AWS, and GCP integration. Leverage the best LLMs from our research.

Learn More

Who We Serve

LearnLens adapts to diverse educational environments, delivering tailored solutions for every learning context.

K-12 Education

Empower teachers with AI-assisted grading and provide students with personalized feedback to enhance learning outcomes in primary and secondary schools.

Higher Education

Scale assessment capabilities for universities and colleges with automated grading, plagiarism-resistant evaluations, and detailed learning analytics.

Corporate Training

Accelerate employee development with intelligent assessment tools that track skill acquisition and provide targeted recommendations for professional growth.

Exam Boards

Support examination bodies with scalable, consistent, and explainable automated marking solutions that maintain assessment standards and reliability.

K-12 Education

Empower teachers with AI-assisted grading and provide students with personalized feedback to enhance learning outcomes in primary and secondary schools.

Higher Education

Scale assessment capabilities for universities and colleges with automated grading, plagiarism-resistant evaluations, and detailed learning analytics.

Corporate Training

Accelerate employee development with intelligent assessment tools that track skill acquisition and provide targeted recommendations for professional growth.

Exam Boards

Support examination bodies with scalable, consistent, and explainable automated marking solutions that maintain assessment standards and reliability.

Watch In Action

See how LearnLens transforms the educational experience with our demo videos and tutorials.

AERAChat Platform

Automated assessment platform with multiple LLM supported.

LearnLens Lite Platform

Personalized learning platform for class use.

Meet Our Team

World-class researchers and engineers dedicated to transforming education through AI.

Jiazheng Li

Jiazheng Li

Co-Founder and CEO

AQA-funded PhD Researcher at King’s College London, focused on educational technology and AI in education, with extensive research experience in industry labs.

Prof. Yulan He

Prof. Yulan He

Co-Founder and Chief Scientific Advisor

A UKRI-funded Turing AI Fellow and Professor of Natural Language Processing at King’s College London, focused on LLM reasoning, agentic AI, long-context QA, interpretability, and safety alignment, with award-winning work and a PhD from University of Cambridge.

Dr. Runcong Zhao

Dr. Runcong Zhao

Co-Founder and CTO

Senior Research Scientist in Large Language Models at King’s College London focused on NLP and machine learning for LLM reasoning and interactive AI applications, with a PhD in Computer Science (University of Warwick) and a BA in Mathematics (University of Cambridge).

Artem Bobrov

Artem Bobrov

Co-Founder and Founding Engineer

Robin Zhou

Robin Zhou

Founding Cloud Architect

Fangkang Mi

Fangkang Mi

Founding Full Stack Engineer

Harry Li

Harry Li

Founding Full Stack Engineer

Chengyu Cao

Chengyu Cao

Research Scientist

Join Our Team

We're always looking for passionate people to help transform education.

View Open Positions

Our Research

Peer-reviewed publications driving the future of AI-powered education.

EMNLP 2025LLM ReasoningSelf-Reflection

Two Heads Are Better Than One: Dual-Model Verbal Reflection at Inference-Time

J. Li, Y. Zhou, J. Lu, G. Tyen, L. Gui, C. Aloisi and Y. He

A dual-model reasoning framework within a verbal reinforcement learning paradigm, decoupling inference-time self-reflection into specialized models for reasoning critique and refinement.

EMNLP 2025PersonalizationFeedback

LearnLens: LLM-Enabled Personalised, Curriculum-Grounded Feedback with Educators in the Loop

R. Zhao, A. Bobrov, J. Li and Y. He

A modular LLM-based system that generates personalised, curriculum-aligned feedback in science education with error-aware assessment and educator-in-the-loop interface.

EMNLP 2025AssessmentExplainability

AERA Chat: An Interactive Platform for Automated Explainable Student Answer Assessment

J. Li, A. Bobrov, R. Zhao, C. Aloisi and Y. He

An interactive visualization platform leveraging multiple LLMs to concurrently score student answers and generate explanatory rationales with intuitive annotation tools.

ACL 2025FaithfulnessRationale

Drift: Enhancing LLM Faithfulness in Rationale Generation via Dual-Reward Probabilistic Inference

J. Li, H. Yan and Y. He

A probabilistic inference paradigm leveraging task-specific and lookahead rewards to ensure LLM-generated rationales are more faithful to model decisions and align better with input context.

AAAI 2025AssessmentLLM

An Automated Explainable Educational Assessment System Built on LLMs

J. Li, A. Bobrov, D. West, C. Aloisi and Y. He

An automated and explainable educational assessment system designed for interactive and visual evaluations of student responses using microservices architecture.

EMNLP 2024Preference OptimizationReasoning

Calibrating LLMs with Preference Optimization on Thought Trees for Generating Rationale in Science Question Scoring

J. Li, H. Xu, Z. Sun, Y. Zhou, D. West, C. Aloisi and Y. He

A framework achieving 38% improvement in QWK score by generating thought trees and utilizing synthetic data for supervised fine-tuning and preference optimization.

EMNLP 2023ChatGPTDistillation

Distilling ChatGPT for Explainable Automated Student Answer Assessment

J. Li, L. Gui, Y. Zhou, D. West, C. Aloisi and Y. He

A novel framework exploring ChatGPT for concurrent student answer scoring and rationale generation, improving QWK score by 11% compared to ChatGPT.

BERA 2023InterpretabilityScoring

Standing on the Shoulders of ChatGPT: Towards the Development of an Interpretable Automated Scoring System

J. Li, L. Gui, D. West, C. Aloisi and Y. He

Pioneering work on interpretable automated scoring systems leveraging large language models for educational assessment.

In the News

See what the media is saying about LearnLens and our mission to transform education.

The TelegraphJuly 12, 2025

AI to Help Mark Student Exams

AQA and King's College London receive over £600,000 to create an AI virtual assistant that adds quality control for exam markers. "The goal is not to replace examiners, but to support them to mark more consistently and faster," says Prof Yulan He.

Read Article
King's College LondonJanuary 20, 2025

King's to Build AI Tool for Teachers as Part of Government's AI Drive

King's College London is developing an AI assessment platform for secondary school science education, delivering personalized feedback while reducing teacher workload through automated marking.

Read Article
King's College London2025

Meet the Spinouts on King's Spinout Accelerator Cohort II

LearnLens (formerly X-Mark) joins King's Spinout Accelerator — an AI-driven assessment platform providing explainable grading and personalized feedback for short essay questions, led by Prof Yulan He, Dr Runcong Zhao, and Jiazheng Li.

Read Article
UKRI2025

UK Businesses and Academia Partner Up in Cutting-Edge Research

King's College London and AQA receive £41 million EPSRC funding to develop AI-enabled high stakes assessment, creating a safe and reliable virtual assistant to support examiners.

Read Article
The TelegraphJuly 12, 2025

AI to Help Mark Student Exams

AQA and King's College London receive over £600,000 to create an AI virtual assistant that adds quality control for exam markers. "The goal is not to replace examiners, but to support them to mark more consistently and faster," says Prof Yulan He.

Read Article
King's College LondonJanuary 20, 2025

King's to Build AI Tool for Teachers as Part of Government's AI Drive

King's College London is developing an AI assessment platform for secondary school science education, delivering personalized feedback while reducing teacher workload through automated marking.

Read Article
King's College London2025

Meet the Spinouts on King's Spinout Accelerator Cohort II

LearnLens (formerly X-Mark) joins King's Spinout Accelerator — an AI-driven assessment platform providing explainable grading and personalized feedback for short essay questions, led by Prof Yulan He, Dr Runcong Zhao, and Jiazheng Li.

Read Article
UKRI2025

UK Businesses and Academia Partner Up in Cutting-Edge Research

King's College London and AQA receive £41 million EPSRC funding to develop AI-enabled high stakes assessment, creating a safe and reliable virtual assistant to support examiners.

Read Article